# Ch 6 Practice — Pipeline Coverage by Region

> The six-region pipeline dataset behind Chapter 6's *Practice with us* problem in *The Defensible Decision*. Six series, a five-colour palette, and four presentation failures stacked on one chart — each of which, on its own, would be survivable.

**Wholly synthetic.** Any resemblance to a specific company's pipeline is coincidental. The values are constructed so that three regions decline while only one falls below the threshold that matters, which is the condition the exercise depends on.

## Files

- [`practice-ch6-pipeline-coverage.csv`](practice-ch6-pipeline-coverage.csv) — tidy long rows: six regions × four quarters, with the palette slot each region was assigned
- [`practice-ch6-pipeline-coverage.json`](practice-ch6-pipeline-coverage.json) — canonical scenario contract, including all four critique surfaces and their Refine directions
- [`SCHEMA.md`](SCHEMA.md) — column definitions, headline numbers, design rationale

## What it represents

A pipeline-coverage-by-region line chart of the kind Copilot will generate without being asked twice. Six regions, four quarters, coverage stated as a percentage of quota.

Three regions decline across the window: North America, ANZ, and EMEA. Only EMEA is below `100%`, and it closes at `80%`. Coverage under `100%` means the pipeline no longer covers quota at all, so EMEA's decline is the only one that changes what anyone should do next. That distinction is the Big Idea, and the chart as generated does nothing to carry it.

## The four failures

Chapter 6 shows the chart carrying four discipline failures simultaneously. Each is individually recoverable; together they make the Big Idea unreachable.

| Surface | What went wrong |
| --- | --- |
| **Palette type** | A five-colour categorical brand palette is asked to carry six series. ANZ falls back to gray — and reads as de-emphasised despite carrying the highest values. |
| **Accessibility** | APAC (`#0d9488` teal) and LATAM (`#15803d` green) collapse together under deuteranopia. They open only ten points apart, so position does not rescue them either. |
| **Typographic hierarchy** | Chart title, axis title, tick labels and end-of-line values are all set at the same size and weight. Nothing signals what to read first. |
| **Pre-attentive attributes** | All six lines share one stroke weight and one saturation level. No pre-attentive channel is doing any work, so the eye ranks nothing and scans linearly. |

## The exercise

Name each of the four failures, then give the Refine direction that corrects it. The JSON records the reference answers under `four_failures`, each with a `failure` description and a `refine_direction`.

The harder half of the exercise is the ordering. The four fixes are not independent: choosing direct end-of-line labels to solve the accessibility problem also removes the legend, which changes what the typographic hierarchy has to do. Deciding which failure to correct *first* is the judgement the chapter is teaching.

## Invariants (locked)

These are pinned by `practice-ch6-pipeline-coverage-contract.test.mjs` in the book repository, so any chart built from this data reproduces the print figure exactly:

- EMEA runs `95%` to `80%`, and is the only region below `100%`
- ANZ closes at `155%`, tied with APAC — the collision the end-of-line labels have to resolve
- Exactly three regions decline, so decline alone does not isolate the story
- Every series moves monotonically, so no region is noisy enough to distract
- Six series against five brand slots, with exactly one gray fallback — if this ever stops being true, the palette critique no longer has anything to point at
- All four critique surfaces are present, each with a Refine direction

## Scope

A practice-scope dataset, not one of the fourteen character datasets. No anchor character owns it; it belongs to the chapter's practice tier. It pairs with [`practice-ch5-revenue-plan`](../practice-ch5-revenue-plan/), the Chapter 5 practice dataset: Chapter 5 asks what to *remove* from a chart, Chapter 6 asks what the remaining marks should *look like*.

## License

Released for teaching use alongside *The Defensible Decision*. Use freely.